The configuration in the peptide compound represented by the foregoing formula (8) influences the activity. In particular, it has been ascertained that the preferred configuration at the two asymmetric carbon atoms in the carboxylic acid ester moiety is of a (2R,3S)-configuration.
It has, therefore, been desired to develop a process for producing, in an industrially advantageous manner, an optically active (2R,3S)-cyclohexylnorstatine represented by the foregoing formula (9), which constitutes an important asymmetric moiety of the above-described peptide compound.
Hitherto, as a process for producing (2R,3S)-cyclohexylnorstatine, a process in which phenylalanine is used as a raw material is reported in J. Chem. Soc., Chem. Commun., 1989, p. 1678, Chem. Lett., 1990, p. 723, J. Med. Chem., 1990, 33, p. 2707, and JP-A-1-172365. In this process, an alcohol derived from phenylalanine is oxidized to an aldehyde, which has then added thereto prussic acid gas or the like to form two optically active sites. However, this process involves problems in application to commercial production because the process includes the steps of oxidation reaction and of using a toxic cyano compound. In addition, since the aldehyde formed in the course of the process is very unstable, it is likely racemized so that it was extremely difficult to obtain products having a high optical purity.
As another process for producing (2R,3S)-cyclohexylnorstatine, a process for production from a 4-cyclohexylmethyl-2-azetidinone derivative is reported in JP-A-2-121963. However, this process was also not satisfactory in the yield and optical purity.
In addition, as a process for producing optically active (2S,3S)-cyclohexylnorstatine of formula (10): ##STR3## a process for inverting the configuration at the 2-position of the (2R,3S)-isomer is known (J. Med. Chem., 1990, 33, p. 2707).
It is reported in JP-A-63-183551 that Z-4-cyclohexyl-2-buten-1-ol (3C) is epoxidized to 2(S),3(R)-epoxy-4-cyclohexyl-1-butanol (3D), which is then oxidized by ruthenium tetrachloride and periodic acid to obtain 2(S),3(R)-epoxy-4-cyclohexyl-1-butanoic acid (3E').
However, as illustrated in the following reaction scheme, it is considered that compound (3E') described in this Japanese patent is clearly erroneously expressed and should read 2(R),3(R)-epoxy-4-cyclohexyl-1-butanoic acid (3E). This is because while the steric structure at the 2-position is kept by the oxidation of compound (3D), since the carbonyl group is introduced at the 1-position, the steric expression at the 2-position should be (R) but not (S). It is also reported in this Japanese patent that this 2(R),3(R)-epoxy-4-cyclohexyl-1-butanoic acid (3E) is reacted with lithium azide to obtain 3(S)-azido-2(R)-hydroxy-4-cyclohexylbutyric acid (3F). ##STR4##